Blood transfusion has no effect on colorectal cancer survival. A population-based study.

This study was conducted to evaluate the impact on survival of perioperative blood transfusion in a series of 698 colorectal cancer patients undergoing radical surgery. Patients were identified, and follow-up was carried out by the local population-based cancer registry. Data on blood transfusion was obtained by record linkage with the files of the blood banks operating in the area covered by the registry. Prognostic factors were age, Dukes stage and topography of the primary tumour. Relative risk (RR) for Dukes B patients was 1.53 [95% confidence interval (CI) 0.94-2.50] and for Dukes C, 3.57 (95% CI 2.22-5.75) when compared with Dukes A patients. For the left colon, RR was 0.96 (0.61-1.52) and for the rectum 1.87 (1.22-2.86) when compared with the right colon. When adjusting for these factors and excluding operative mortality, RR for transfused patients was 1.16 (95% CI 0.87-1.55). It is concluded that blood transfusion does not adversely affect survival in colorectal cancer patients.